« January 2008 | Main | March 2008 »

A New Hampshire Post-Mortem

February 29, 2008 11:57 AM

Pollsters shed some light on their New Hampshire problem last night, with the Gallup Organization reporting that half the misstatement in its final pre-election poll was caused by its likely voter modeling. But other pollsters differed, agreeing chiefly that the causes of the meltdown remain elusive.

The discussion came at a meeting of the New York Chapter of the American Association for Public Opinion Research, whose national organization separately is looking into the failure. Each of nine polls completed the Sunday or Monday before the Jan. 8 New Hampshire primary showed Barack Obama numerically ahead, by 3 to 13 points, averaging 8. He lost by 2.

Gallup, whose final poll had Obama ahead by 13 points, had a closer 5-point Obama lead among people who described themselves as registered voters. That means its likely voter modeling, used to produce a more accurate estimate of who’ll actually vote, instead introduced error.

Gallup’s editor-in-chief, Frank Newport, said the modeling included factors such as enthusiasm and attention to the race, both of which may have increased for Obama and slacked off for Hillary Clinton after Obama’s Jan. 3 victory in Iowa. Unlikely voters – those excluded from the model – were much better for Clinton. “Obviously that was a cause for the incorrect likely voter numbers that Gallup put out,” he said.

Given Obama’s lead even among self-reported registered voters, Newport suggested two possible additional factors: late changes in voter preferences and a more effective get-out-the-vote effort by the Clinton campaign. (Presumably bias in self-reporting of registration is possible as well.)

Another polling director, Lee Miringoff of the Marist College Institute for Public Opinion, said likely voter modeling was not a factor in his poll, which showed an 8-point Obama lead. Instead, he said, “Our data suggest there was some kind of late shift to Hillary Clinton among women.”

That conclusion runs counter to data from the exit poll, which showed no large swing to Clinton among late deciders; and two final pre-election polls that if anything found an even larger Obama lead Monday night. The rest of the polls, Gallup’s and Marist’s included, were completed Sunday, a shortcoming Miringoff said he will not repeat.

A third panelist, Sarah Dutton of CBS News, whose poll had a 7-point Obama lead, similarly surmised that the New Hampshire polls “picked up a post-Iowa bounce” for Obama that didn’t carry through to Election Day. She said pollsters may have taken false comfort from the fact that all their data were similar, failing to scrutinize warning signs such as greater strength of support on Clinton’s side.

Otherwise, she said, “We might have had a little bit of an alarm bell saying, gee, we have disproportionate strength of support for these candidates.” Dutton said her analysis was limited by the sample size of the CBS poll, 323 respondents reinterviewed from a November survey.

Newport and Miringoff based their conclusions partly on post-election polls in which they called back respondents to their pre-election polls in an effort to see where those polls went wrong. Analysis of those data is not complete, though Newport said Gallup hopes to post some conclusions on its website next week.

Both said their callback polls reached about two-thirds of the original poll respondents; they hadn’t yet weighted these samples to adjust for the noncoverage, a step that could improve their analysis.

Newport said the people he recontacted divided 37-35 percent for Obama over Clinton in the pre-election poll, while reporting that they actually voted for Clinton over Obama by 39-37 percent, suggesting some changed their minds after the pre-election poll was done. At the same time, he noted the hazards of recall and self-reporting errors in post-election polls.

Marist’s callback poll found a self-reported vote of 37-31 percent in Clinton’s favor (a wider lead than Gallup’s, and than the actual result) compared with its pre-election poll (among all respondents, not just those recontacted on callback) of 36-28 percent for Obama. Twenty percent said they’d changed their minds, of whom nearly half went to Clinton vs. two in 10 for Obama, supporting the notion of a late shift in her favor, he said.

However, the New Hampshire exit poll asked people when they’d “finally decided” whom to support; 17 percent said it’d been in the last day, and they split 39-36 percent for Clinton, not nearly enough to account for the Obama overstatement in the polls. An additional 21 percent said they’d decided in the last three days, and they split 37-34 percent for Obama, again insufficient to explain the pre-election polls.

Dutton suggested that the exit poll question may have been difficult to answer accurately for people who’d switched their preference once or more.

Newport said eight in 10 voters contacted in his callback poll reported having seen the video clip in which Clinton became emotional the day before the election, and more said it made them react positively toward her than negatively. “The video may have had a positive impact,” he said. (He told a questioner that the poll didn’t ask respondents when they’d seen the video – before or after the election, when it received frequent replay.)

Miringoff, similarly, reported that eight in 10 had seen the video, and women in particular reacted to it more positively (31 percent) than negatively (11 percent).

Newport also said Clinton supporters also were more apt to say they’d been contacted by her campaign and offered assistance getting to their polling places, a possible measure of get-out-the-vote effectiveness.

In the future, Miringoff said he’d poll through the Monday before Election Day, and Newport said Gallup would re-evaluate the likely voter model it used in New Hampshire. Both also pledged further data analysis; so far, “What we have are inferential or suggestive data. None of these are huge numbers,” Newport said. “There’s no smoking gun.”

February 29, 2008 | Permalink | User Comments (29) | TrackBack (0)

+16, +12, +3, +2

February 26, 2008 10:38 AM

The New York Times this morning reported two national polls completed Sunday that show a double-digit lead for Barack Obama over Hillary Clinton in preference for the Democratic nomination - the Times' own, with CBS News; and one from Gallup with USA Today. There were, however, two others - from Gallup alone and from AP/Ipsos - that show a much closer race, virtually dead heats (Obama +2 and +3).

With these differences beyond sampling error, the reason why is a puzzlement. Frank Newport of Gallup has a piece on Gallup.com discussing how his organization has two polls on the same day, one with Obama +12, the other with Obama +2, but he arrives at no firm conclusions. Question order always is a possible culprit, but it doesn't usually make for differences like these. All four polls had about the same numer of undecideds. One was among likely voters, but the other three were among the general population. And the two conflicting Gallup polls had the largest sample sizes.

I’ve noted before that primary polling, lacking the anchoring influence of partisanship, can be a squirrelly affair. Sampling approaches can cause differences (e.g., ABC News and The Washington Post alone have oversampled African-Americans in every poll we've done this cycle). And there can be flexion points around which polls simply differ because attitudes are unsettled.

For the moment, with the cause of these differing estimates up in the air, when considering the two national polls that show Obama ahead it would be prudent also to keep in mind the two that show the race essentially tied.

But all of them, of course, are a far cry from the substantial Clinton lead we saw in ancient times - like a month ago.

Source          End date  Interviews  Pop. Clinton  Obama   Diff.
CBS/NYT          2/24/08      427     LV      38      54     +16   
AP/Ipsos         2/24/08      473     GP      43      46      +3
Gallup           2/24/08    1,294     GP      45      47      +2
USA Today/Gallup 2/24/08    1,009     GP      39      51     +12  

February 26, 2008 | Permalink | User Comments (54) | TrackBack (0)

Spoilage?

February 25, 2008 11:28 AM

Ralph Nader’s announcement of his latest tilt at the presidency promptly launched a fresh round of the parlor game called “spoiler” – this round arguing that Nader could cost the Democratic nominee the presidency in 2008, as he allegedly did in 2000.

Allegedly is the operative word. Whether Nader indeed cost Al Gore the presidency is less of an open-and-shut case than you might think. And whether we’ll see another storm so perfect that a 2.7 percent candidate can be even accused of tipping the balance is hardly a sure bet.

The spoiler claim goes like this: George W. Bush won Florida in 2000 by 537 votes. Nader had 97,488 there. We can’t be sure how Nader’s Florida supporters would have voted had he not been on the ballot, but in national exit poll data (necessary for a sufficient sample size), 47 percent said they’d have voted for Gore, 21 percent for Bush, and the rest would’ve stayed home. Divide the Nader vote in Florida that way, and inaugurate President Gore.

But wait. If you’re going to say Nader cost Gore a win in Florida, you might have to say the same thing about any of a list of lesser-known possible spoilers. David McReynolds, the Socialist candidate in 2000, got 622 votes in Florida. Give ’em to Gore and he’d have won regardless of Nader. Or give Gore the lion’s share of the Libertarian, Constitution Party or Natural Law Party vote in Florida, and get there that way.

But don't forget Pat Buchanan, the other free-lancer in 2000. He didn’t get enough votes to estimate where they’d have gone otherwise, but if you give all or most of them to Bush, you can do some serious theoretical mayhem. Buchanan got 17,484 votes right there in the Land of the Hanging Chad. And look beyond: In New Mexico, where Gore won by 366 votes, Buchanan got 1,392. In Iowa, Gore won by 4,144; Buchanan got 5,731. In Oregon, Gore won by 6,765, while Buchanan got 7,063 votes. And in Wisconsin, Gore won by 5,708 but Buchanan got 11,471. Give the lion’s share of those Buchanan votes to Bush and the Nader-in-Florida argument just might become irrelevant.

If this isn’t enough fun, start apportioning out the rest of the 2000 vote beyond Florida. Nationally, candidates other than Gore, Bush, Nader and Buchanan got a total of 593,078 votes. Out of whose hide? Go figure. But while you're at it don’t forget that Gore lost his own home state by more than 50,000 votes, raising the entirely plausible suggestion that it was Al Gore who cost Al Gore the 2000 election.

Part of the fascination with third-party candidacies (cue Mike Bloomberg) comes from farther back, 1992, when it’s alleged that Ross Perot tilted the election to Bill Clinton by swiping votes disproportionately from George H.W. Bush. In fact the national exit poll that year found that had Perot not been in the race his supporters would have divided evenly between Clinton and Bush, at 40 percent each; the rest would have supported someone else, or sat it out.

Perot got 19 percent of the national vote in 1992, the second-best showing by any independent in modern times. Nader’s been a different story. In 1996 he got seven-tenths of one percent; in 2000, without Perot as an alternative independent candidate, 2.7 percent; and in 2004 Nader bottomed out at just under four-tenths of one percent, 465,650 votes out of 122,295,345 cast.

With numbers like these, Nader could end up looking less like a spoiler and more like Harold Stassen, who ran unsuccessfully for the Republican presidential nomination in 1948, 1952, 1964, 1968, 1976, 1980, 1984, 1988 and 1992. But the math on that is tough in its own way: The day after tomorrow, Nader turns 74 years old.

February 25, 2008 | Permalink | User Comments (32) | TrackBack (0)

The SES Factor

February 19, 2008 10:00 AM

My last couple of items have looked at race and sex as factors in the Democratic presidential campaign (and separately we've examined the role of ideology and religious belief on the Republican side). The Clinton-Obama contest includes another crucial variable to watch: socioeconomic status.

With remarkable consistency, Hillary Clinton has done better with lower-education and -income groups, Barack Obama with higher ones. It may matter in Wisconsin today, in Texas, and perhaps more than anywhere else, in Ohio on March 4.

It shows up best when we look at vote preferences by educational attainment. Overall, combining all primaries to date, voters who hold a college degree have voted for Barack Obama over Hillary Clinton by an 8-point margin, 51-43 percent, while those who haven’t been graduated from college have favored Clinton by 10 points.

Breaking it down by sex sharpens the picture: Obama easily has won college-educated men, while Clinton has just about as easily won women who’ve not been to college. The battle’s among two other groups – less-educated men and more-educated women, both of which have divided evenly between Clinton and Obama in primaries so far. (It’s a point on which Ron Brownstein, political editor of the National Journal, has been justifiably fixated all season.)

It gets even more interesting when we take the next step and look at race as a factor as well. Obama’s been winning more-educated white men by a 17-point margin, and losing less-educated white men by 19 points – a dramatic 36-point education gap. Clinton’s been winning more-educated white women by 13 points, while winning less-educated white women by a whopping 38 points – not quite as big an education gap, at 25 points, but hefty nonetheless.

As a result, among whites overall, to date Clinton’s won the less-educated by 30 points, while the college-educated have split evenly.

There’s also an education gap among Hispanics, men and women alike. But there’s been much less of a division among African-Americans, given their very broad overall support for Obama. The tables below tell the story:

             All          All men      All women
         College  No    College  No    College  No
Clinton    43%    51      36     46      48     54
Obama      51     41      57     46      47     39


          All whites     White men     White women
         College  No    College  No    College  No
Clinton    47     59      37     52      53     64
Obama      46     29      54     33      40     26


          All blacks     Black men    Black women
         College  No    College  No    College  No
Clinton     14    18      14     15      14     20
Obama       81    78      82     81      81     75
 
         All Hispanics  Hispanic men  Hispanic women
         College  No    College  No    College  No 
Clinton     55    65      48     59      59     69
Obama       42    32      47     38      39     28


         College/No college difference
               Clinton   Obama
All              -8      +10
All whites      -12      +17
All blacks       -4       +3
All Hispanics   -10      +10
White men       -15      +21
White women     -11      +14
Black men        -1       +1
Black women      -6       +6
Hispanic men    -11       +9
Hispanic women  -10      +11

One conclusion is that Obama has run so competitively overall in part because the primaries attract better-educated people. Among all Democratic primary voters, 48 percent have been college graduates; among whites, 54 percent. Compare that with all Democrats and Democratic-leaning independents in the country: In aggregate 2007 data from ABC News/Washington Post polls, just 29 percent held college degrees. The disproportionate participation of better-educated voters in the primaries has paid off for Obama big-time.

And looking ahead? In Democratic primary exit polls from 2004, 46 percent of whites in Wisconsin and Texas alike held college degrees; in Ohio fewer did, 39 percent. Those figures don’t predict who’ll turn out this year. But it’s clear that along with race and sex, the education level of Democrats who show up to vote will be a critical factor in the outcome of these primaries.

February 19, 2008 | Permalink | User Comments (16) | TrackBack (0)

The Role of Race

February 15, 2008 9:35 AM

Race has been a riveting factor in the Democratic presidential primaries; even beyond sex, age and socioeconomics, it looks to be the single most powerful demographic in vote choices – at least for nonwhites.

Witness New Mexico, which Hillary Clinton last night was announced to have won by a razor-thin 1,709 votes – despite losing white voters there by a 12-point margin. The reason: her 26-point victory among Hispanics.

I wrote last week about white men as a swing group; it proved out in Virginia and Maryland. But differing vote preferences between African-Americans and Hispanics are essential as well, and especially worth evaluating with an eye toward the Texas primary ahead.

There is some tension here. In aggregate exit poll data from the Super Tuesday states, just 3 percent of whites called the race of the candidate “the single most important factor” in their vote; that rose to 12 percent among Hispanics. Voters in both those groups favored Clinton by roughly 3-1 margins, 73-22 percent among whites, 72-27 percent among Hispanics – in both cases, better for Clinton than she did among whites and Hispanics who gave less importance to the candidates’ race.

Among black voters, 6 percent called race the single top factor and 83 percent of them went for Obama, but so did 82 percent of all other blacks.

Obama surprised Clinton among Hispanics in Virginia this week, seemingly beating her by 54-46 percent; that only happened previously in Connecticut, 53-43 percent. In fact, though, the sample size in both states was too small for reliable analysis, and the difference between the candidates was within sampling tolerances.

More reliable is the overall result among Hispanics: Across all primaries to date, Clinton’s won them by 61-34 percent. Her 67 percent support from Hispanics in California (where Obama won white men by 20 points) and by 55 percent in Arizona (where white men split evenly) were crucial to her winning those states. New Mexico’s the latest example.

As with whites, socioeconomic status plays into the Hispanic vote. Clinton has done 10 points better among Hispanics who don’t have a college degree (65-32 percent, vs. 55-42 percent among Hispanic college graduates) and among those with less than $50,000 in family incomes (67-30, vs. 57-40 percent among better-off Hispanics). She’s helped in this population by the fact that Hispanic voters are less likely to be college graduates (32 percent vs. 39 percent of blacks and 54 percent of whites) and more apt than whites to have incomes under $50,000 (49 percent of Hispanics vs. 33 percent of whites.)

Clinton’s done 11 points better with Hispanic women (66-31 percent over Obama) than with Hispanic men (55-39 percent). Among whites, she's won women overall by 59-34 percent, but managed only a dead heat among men, 45-44 percent.

Among blacks, meanwhile, the surge to Obama since December has been remarkable. In the last pre-primary ABC News/Washington Post poll last year, Clinton led Obama among blacks by 52-39 percent. That changed after Obama established his credentials by winning Iowa; across all primaries to date he’s won African-Americans by 79-17 percent.

Whites have accounted for 61 percent of Democratic primary voters; 20 percent have been blacks, ranging from a low of 1 percent in New Hampshire to a high of 55 percent in South Carolina. Hispanics have accounted for 14 percent overall, peaking at 35 percent in New Mexico and 30 percent in California, where their turnout was up sharply from 16 percent in 2004. The Texas primary is March 4; in 2004 Hispanics accounted for 24 percent of Democratic primary voters there, blacks, 21 percent.

All these number stand in sharp contrast to the Republican primaries, in which voters have been divided much more by religious belief (evangelical vs. non-evangelical) and ideology. There’s good reason race hasn’t been a factor: Whites have accounted for 88 percent of Republican primary voters, compared with 61 percent of Democrats. Hispanics, at 6 percent of GOP voters, are less than half as prevalent in the Republican ranks as in the Democratic. And blacks, two in 10 Democratic voters, account for just 3 percent of Republicans in this year's primaries.

February 15, 2008 | Permalink | User Comments (80) | TrackBack (0)

In the Democratic Race, Whither White Men

February 11, 2008 12:26 PM

Tomorrow’s primaries may present an opportunity to look at the preference of white men for Obama over Clinton in some contests. While far from consistent in primaries, it’s been an important factor in their support profiles: Where white men favor Obama, he's far likelier to win.

Mason Dixon’s polling has Obama winning 41 percent of whites in Virginia and 40 percent in Maryland; that’s essential to his overall lead in these two states, since in neither of them are African-Americans predicted to have a large enough share to carry Obama themselves (as in Louisiana on Saturday, and in some previous Southern states). That suggests Obama could win white men on Tuesday, as he has in seven of the 23 previous states for which we have data. (Clinton's won them in 11, and they've tied in five.)

In some states to date where Clinton has not won white men – California and Arizona – there’ve been enough Hispanics to make the difference for her. In some others (New Hampshire, Massachusetts) there have been few blacks, and enough white women to outweigh white men. But in states with significant but not vast numbers of black voters, and few Hispanics, white men are critical. Per the table below, Clinton’s won the primary in eight of the 11 states where she won white men, a 73 percent success rate; but in only four of the 12 where she didn’t win them, 33 percent.

When we look at issue preferences, we find white men generally less attuned to Clinton across a range of parameters. In aggregate data, white men have been 17 points less apt than white women to say Clinton would be the best commander-in-chief; 17 points less apt to call "experience" the most important attribute (it's a strong Clinton quality), 17 points less apt to say Clinton would do the most to unite the country and 14 percent less apt to call health care their top issue (also strong for Clinton).

Naturally, affinity voting plays a major role. White women are twice as apt as white men to call the sex of the candidate important in their vote, 35 percent to 17 percent. And white women who call the candidate's sex important have favored Clinton over Obama by a huge 78-17 percent margin; white men who call it important also have favored Clinton, but by a much closer margin, 54-35 percent.

Addendum: With thanks to Ron Brownstein of the National Journal, there’s a further point of differentiation here – the additional role of socioeconomic status. Education provides the clearest view: Combining all primaries to date, white men who’ve attended college have gone to Obama over Clinton by 51-38 percent, while white men without a degree have gone the other way – 52-31 percent for Clinton; he does 20 points better, she 14 points worse, among college educated white men vs. non-graduates. There’s a similar division among white women, albeit a little less pronounced – among those with college degrees, Obama is +14 points, Clinton -10, compared with those who haven’t earned a degree. There’s less of an education gap among nonwhites.

Table's below. One note – apart from Iowa and Nevada we have no data for caucuses; just primaries. Caucuses, with their very low turnout, can be something of a different animal.

            Vote preference 
            among white men
         Clinton Obama Edwards  Winner
Clinton:
Arkansas   71%    24            Clinton
Alabama    70     27            Obama 
New Jersey 58     39            Clinton
Tennessee  58     32            Clinton
Oklahoma   55     32            Clinton
Missouri   55     41            NA
Michigan   54     38*           Clinton
New York   52     43            Clinton
Louisiana  51     31            Obama
Florida    45     27     25     Clinton
Nevada     46     40     10     Clinton


Tied:
Mass.      48     49            Clinton
Delaware   47     47            Obama
Arizona    46     45            Clinton
Georgia    46     48            Obama
S.C.       28     27     45     Obama


Obama:
Utah      29      64            Obama
Illinois  37      59            Obama
N.M.      38      59            NA
Conn.     40      57            Obama
Calif.    35      55            Clinton
N.H.      30      38     20     Clinton
Iowa      23      33     25     Obama



*uncommitted

February 11, 2008 | Permalink | User Comments (71) | TrackBack (0)

Turnout and the Youth Vote

February 08, 2008 11:12 AM

This note takes a look at turnout overall in primaries to date, and then tackles the question of turnout among young voters. In short, with the combined punch of an exciting race and earlier dates, turnout in Democratic races is up. Turnout among young voters, though, is a more complicated story – up in some Democratic races, but not in others.

TURNOUT OVERALL – It’s risen, especially in Democratic contests, and especially this week. Turnout on Tuesday peaked at 28 percent of eligible voters in the Massachusetts Democratic primary, more than double its 2004 level; 23 percent in Illinois, up from 14; and 20 percent in New Jersey, up from just 4 percent in the state’s June 2004 primary.

Turnout in Republican races was much lower, and generally did not increase sharply. Compared to 2000, it was up by 10 points in Alabama and Utah, and by 8 in Oklahoma and Arkansas, but down by 12 in New York and by 4 points in California.

All this is tabled below; figures are as a percentage of eligible voters, and reflect votes reported to date – there still are some absentee votes to add. Per AP's vote count, the totals to date are 18,984,677 in Democratic presidential primaries and caucuses; 12,862,478 in the Republican races.

               Dem. Turnout         Rep. Turnout
                     % of eligible voters
             2008   2004 (date)   2008    2000 (date)
Super Tues - Primaries:
Mass.         28    13 (3/2)       11     11 (3/7)
Illinois      23    14 (3/16)      10      9 (3/21)
New Jersey    20     4 (6/8)       10      4 (6/6)
California    19    15 (3/2)       11     15 (3/7)
Missouri      19    10 (2/3)       14     12 (3/7)
Alabama       16     7 (6/1)       16      6 (8/6) 
Georgia       17    11 (3/2)       15     12 (3/7)
Oklahoma      16    12 (2/3)       13      5 (3/14)
Delaware      16     6 (2/3)        8      6 (2/8)
Arkansas      14    13 (5/18)      10      2 (5/23)
Connecticut   14     5 (3/2)        6      8 (3/7)
New York      14     6 (3/2)        5     17 (3/7)
Tennessee     14     9 (2/10)      12      6 (3/14)
New Mexico    11    NA             NA
Arizona       10     7 (3/3)       12     10 (2/22)
Utah           7     2 (2/27)      16      6 (3/10)

Super Tues – Caucuses:
Colorado       4    NA              2      6 (3/10)
Idaho          2     3 (5/25)      NA
Kansas         2    NA             NA
Minnesota      6    NA              2      1 (3/7)
North Dakota   4    NA              2      2 (2/29)
Wyoming       NA                    0.3    NA

Previous:  
Iowa          11     6 (1/19)       5      4 (1/24)
New Hampshire 29    23 (1/27)      24     27 (2/1)
Michigan       8    NA             12     18 (2/22)
Nevada         7    NA              3      9 (3/21) 
S.C.          17     9 (2/3)       14     20 (2/19)
Florida       14     6 (3/9)       15      7 (3/14)

YOUNG VOTERS – This leads us to the question of turnout among young voters. It’s gotten a lot of buzz since the Democratic race in Iowa, where young voters did turn out in disproportionately greater numbers, accounting for 22 percent of the party’s caucus-goers. But the story since then has been inconsistent, and Iowa remains the high-water mark.

The analytical challenge is that there are many ways to slice the pie. Overall, using what seems to us a reasonable approach, we’d say that on a state-by-state basis, turnout among voters under 30 is up a little in Democratic contests, and flat in Republican contests.

In probably the simplest way to do this, in all 2008 Democratic contests to date, 18- to 29-year-olds have accounted for 14 percent of voters. That compares to an average of 12 percent in 1992, for states for which we have exit poll data – 2 points higher this year. (It was lower in 2004, 9 percent, and 2000, 8 percent, but looks to have been higher in the 1980s.)

In Republican contests, young-voter turnout is 11 percent this year, vs. 12 percent in 1992 (and 10 percent in 1996, 9 percent in 2000).

 Under 30s as % of all voters
           Dem   Rep
    2008   14    11
    2004    9    NA
    2000    8     9
    1996   NA    10
    1992   12    12

Those are averages. A state-by-state evaluation shows a mixed pattern. All told, there are nine states in which Democratic turnout among young voters has been higher this year than its previous high going back to 1992 (by 2 or more points), eight flat and one down, with no data for the rest.

In Republican races, there are two in which young voter turnout has been up, six in which it’s been down, 10 flat, with no data for the rest.

Among some of the notable increases in Democratic races, young voters accounted for 16 percent of voters in California this week, up from 12 percent in 2000 and 11 percent in 2004 (but no higher than in 1984). It was 18 percent in Georgia, compared with a previous high (since 1992) of 15 percent. And it was 14 percent in South Carolina, compared with 9 percent in 2004.

Note, this looks at turnout among young voters as a percentage of all voters. In all those states where turnout was up overall, young people increased their turnout along with everybody else; the point is that while in some cases they did so disproportionately, in others they did not.

Here’s a list of turnout by 18- to 29-year-olds by state to date, with their previous highs in exit polls since 1992:

            Under 30s as % of all voters     Change
              Dem   Pvs.*   Rep   Pvs.*     Dem   Rep 
  Iowa        22    17      11    13        Up    Down
  Georgia     18    15      11    12        Up    Flat
  N.H.        18    17      14    15        Flat  Flat
  Michigan    17    14      13    13        Up    Flat 
  Utah        17    NA      16     9        NA    Up
  California  16    12      10    12        Up    Down
  Illinois    15    14      10    10        Flat  Flat
  New York    15    10       9     9        Up    Flat
  Missouri    14     9      13    10        Up    Up
  Mass.       14    16      13    20        Down  Down
  New Jersey  13     8       8     8        Up    Flat
  Alabama     13    13      12    11        Flat  Flat
  Tennessee   13     9      11    13        Up    Down
  S.C.        14     9      10    10        Up    Flat
  Nevada      13    NA      11    NA        NA    NA
  Delaware    10     9      NA    NA        Flat  NA    
  Conn.       10    11      11    10        Flat  Flat
  Florida      9     8       7    10        Flat  Down
  Oklahoma     9    10      14    15        Flat  Flat
  Arkansas     9    NA      10    NA        NA    NA
  New Mexico   8    NA      NA    NA        NA    NA
  Arizona      8     7       6     9        Flat  Down


*Previous high since 1992

An interesting aside is that while the youth vote is most closely associated with Barack Obama – he won under-30s nationally by 16 points, 57-41 percent – this was not the case in every state. In Arkansas and Oklahoma, Hillary Clinton won under 30s; in California and Massachusetts, she and Obama split them evenly. Obama has a clear lead in this group – but not a lock.

February 8, 2008 | Permalink | User Comments (20) | TrackBack (0)